Note carefully that this was a Layer 2 test! If this
was a layer 3 test the 6509's throughput would be even
lower. Compare this to Nortel's Passport 8600 switch
which will pass 96 million packets/sec in L2 or L3. I
would think that the throughput of the switches in the
core of most companies networks would be considered
under the heading of technical needs. At this price
point it is important to get what you're paying for in
terms of performance.

Meanwhile Cisco has the gall to claim 256 GB/sec
throughput for the 6509 which is actually the sum
total of the aggregate I/O of all the blades in a
fully populated box multiplied by a factor of two 
(on ingress and egress).

Somehow people still don't seem to get it. I would be
ashamed to sell Catalysts to a customer knowing that
when they got it all in and the network performance
still s*cked it could mean his job.

Caveat Emptor...let the buyer beware !

--- John Hardman  wrote:
> Hi
> 
> It's true. It's also true that in similar tests with
> a Foundry will also out
> perform a Cat. But keep in mind that a lot of this
> works out to be FUD.
> Sales people from each company will have various
> reasons why you should
> choose their product over the other. The bottom line
> is that you have to
> choose which is right for your company based on it's
> business and technical
> needs.
> 
> Both Extreme and Foundry are making a strong push
> into Cisco's enterprise
> switch market share. Their products are very
> competitive, especially at the
> price point. If I could get switches with Foundry's
> architecture, Extreme's
> network management software and CLI, and Cisco's end
> to end solutions, I
> would be a very happy engineer!
> 
> $0.02
> --
> John Hardman CCNP MCSE
> 
> 
> ""mishaal""  wrote in message
> [EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
> > How true is this?
> > Can anyone throw some light on this report from
> www.zdlabs.com, 70-80%
> > packet loss is rather substantial..hope it's not
> true!
> > thanks
> >
> > From ZDLAbs :
> >
> > " In Layer 2 mode, the Black Diamond and Alpine
> switches forwarded 100%
> > of the traffic offered during the test
> > without dropping a single packet. This resulted in
> a throughput of 57.1
> > million packets/second for the Black
> > Diamond and over 38 million packets/second for the
> Alpine using 64-byte
> > packets. These results represent
> > the maximum throughput possible, given the port
> configurations of the
> > switches.
> > The Cisco Catalyst 6509 lost over 78% of the
> packets offered during the
> > Layer 2 full mesh test at the 64-byte
> > packet size. According to the Catalyst 6509
> documentation, the 6509
> > switch fabric is capable of forwarding
> > 15 million packets/second. This rate is
> substantially less than the 57.1
> > million packets/second offered during
> > our test, which explains the large packet loss."
> >
> > 'The Black Diamond and Alpine switches
> successfully routed 100% of the
> > packets offered (over 5.7 billion
> > 64-byte packets) during the test without dropping
> a single packet. This
> > results in a Layer 3 throughput of
> > over 95.2 million packets/second for the Black
> Diamond and over 47.6
> > million packets/second for the
> > Alpine with 64-byte packets.
> > The Layer 3 full mesh results for the Cisco
> Catalyst 6509 were very
> > similar to the Layer 2 results. The
> > switch dropped a large number of packets at all
> block sizes (86.86% with
> > 64-byte packets). As in the
> > previous tests with Catalyst 6509 we verified that
> the internal switch
> > counters matched the results from
> > the SmartFlow application and that there were no
> packet errors during
> > the test."
[EMAIL PROTECTED]


__________________________________________________
Do You Yahoo!?
Make international calls for as low as $.04/minute with Yahoo! Messenger
http://phonecard.yahoo.com/




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=14051&t=13837
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]

Reply via email to